Electromechanical brakes have been known for some time. U.S. Pat. No. 5,788,023 discloses a disc brake for a vehicle which can be actuated electrically and whose brake linings can be pressed against the brake disc with the aid of an electric motor. The electric motor transmits its actuation force, via a so-called planetary rolling-contact threaded spindle, onto an axially displaceably mounted piston which interacts with the brake lining.
U.S. Pat. No. 5,829,557 discloses another vehicle disc brake which can be actuated electrically and whose brake linings can in turn be pressed against the brake disc by means of an electric motor serving as an actuator. The electric motor comprises a spindle gear mechanism and, by means of a spindle element which can be of different designs, is connected, in the direction of displacement of the brake linings, to an axially displaceable piston which acts on a brake lining. In this patent, there is optional provision for the use of an additional gear mechanism for converting the torque and rotational speed.
A major problem with conventional brakes with an electric actuator is the high actuator force which has to be applied in order to achieve a sufficient braking effect. The necessary high actuator force and the resulting large power demand of the actuator make it necessary to employ very large drive units, usually electric motors, which have large torques, and are also heavy and expensive. The result of this is that electromechanical brakes have, to date, not become widespread as vehicle brakes, for example.
In order to decrease the energy consumption of the brake actuators, so-called self-energizing actuators have been proposed. Early examples of such self-energizing brakes can be found, in U.S. Pat. Nos. 4,653,614, 4,852,699, 4,946,007, 4,974,704, 5,012,901. A self-energizing brake works according to the principle that the braking force amplifies itself. The friction force between the brake linings and the brake disc give rise, with help of a self-energizing mechanism, to increased force against the brake linings and brake disc. This increased force gives, in turn, rise to increased friction force. Hence, it is possible to produce and control large braking forces by applying relatively moderate forces.
The degree of self-energization defines the relation between the applied force and the actual braking force. The self-energization is strongly dependent on the disc/pad friction coefficient. Normally, the variations in the disc/pad friction coefficient are large, and are dependent on, among other factors, the temperature of the disc and/or pad. Variations in disc/pad friction coefficient are even possible within one and the same brake application.
At a specific disc/pad friction coefficient, μinf, the static reinforcement of the self-energizing mechanism is principally infinity. That means that one can produce and control large brake forces by only applying relatively moderate forces. For disc/pad friction coefficients lower than this specific number, the brake is stable, which means that a pushing force has to be applied to produce brake forces. For disc/pad friction coefficients larger than μinf, the self-energized brake instead will become unstable, which means that a pulling force has to be applied to hold the brake at a specific brake force or else uncontrolled braking (i.e., lockup) can occur.
Dimensioning the self-energizing mechanism can dimension the properties of the self-energized brake. There are three main ways to dimension the self-energized brake:                (1) Self-releasing brake. The self-energized brake is dimensioned so that μinf is greater than every arising disc/pad friction coefficient. The brake will be self-releasing for all possible disc/pad friction coefficients. A pushing force has to be applied to accomplish brake forces.        (2) Self-applying brake. The self-energized brake is dimensioned so that μinf is lower than all possible disc/pad friction coefficients. The brake will be self-applying for all possible disc/pad friction coefficients. A pulling force has to be applied to hold the brake at a specific brake force.        (3) Self-releasing or self-applying brake. The self-energized brake is dimensioned so that the disc/pad friction coefficient can be both greater and lower than μinf, hence it can be active in both the self-releasing and the self-applying domains. (A special case is to dimension the brake so that μinf is around the nominal value for the disc/pad friction coefficient. The average degree of self-energization is then maximized under normal conditions.) The brake will be self-releasing when the disc/pad friction coefficients are lower than μinf, and self-applying when the disc/pad friction coefficients are greater than μinf.        
In cases (2) and (3) above, control based on feedback is critically necessary to stabilize the brake. Otherwise the self-energized brake may become unstable, resulting in uncontrollable braking (i.e., lockup). One alternative is to pull back the brake (without feedback) when sensors normally used for the feedback are detected to be not functioning properly. However, in this case, it is no longer possible to maintain any brake functionality. Further, the brake actuator reinforcement varies extensively when the disc/pad friction coefficient changes. Without control, the variations in brake force would be extensive. It is also necessary to achieve the correct commanded torque/force on all brakes on each axle of the vehicle (unless it is specifically desired that they not be the same, for example during ABS). This is important so that the brake torque/force will not become different on the different wheels at the same commanded brake torque/force.
U.S. Pat. No. 6,318,513 discloses an electromechanical brake which includes an arrangement which brings about self-energization of the actuation force generated by the electric actuator. The brake also includes a device for comparing a setpoint value of a frictional force with the actual value of the frictional force, which device, in the event of a deviation of the actual value from the setpoint value, drives the electric actuator to correspondingly increase or reduce the generated actuation force, and thus approximates the actual value to the setpoint value of the frictional force. However, controlling the electric actuator based upon feedback indicative of the frictional force is disadvantageous for a number of reasons.
One such disadvantage of this configuration becomes apparent particularly when a disc has a thickness that varies during a revolution, which is a quite normal case. This leads to a varying measured clamp or frictional force with a frequency proportional to the wheel rotation speed and the amplitude directly proportional to the brake actuator stiffness. A direct clamp or frictional force feedback would try to compensate for the force error caused by the disc thickness variation. At some vehicle speed this will lead to a high energy consumption compared to an open-loop system as in today's pneumatic system where the disc thickness variations would be uncompensated by the control loop.
Another disadvantage relates to the fact that a typical brake caliper includes a brake actuator working against the disc on one side of the caliper arrangement and with a number of sliding pins to equalize the force between this side of the disc and the other where a fixed pad is located. This is typically known as a “floating caliper” design, and is quite common. Due to ageing and corrosion the sliding pin function may be far from ideal, particularly at the end of the life-cycle of the brake actuator. This may lead to higher static friction of the sliding pins which can cause a slip-stick effect on the equalization of the force between the two sides of the disc. The stick-slip effect may lead to an unstable force sensor signal. Experience shows that introducing this type of signal in a feedback loop may cause problems in the control loop, especially when working at the optimum degree of self-energization.
A further disadvantage relates to the fact that when controlling the self-energized brake actuator with a disc/pad friction coefficient close to the friction coefficient when the static reinforcement is infinity, μinf, the backlash in the mechanical transmission may be taken into account in control calculations. In the mechanical link between the controlling force/torque from the motor to the clamping force of the pad/disc it is difficult and costly to totally eliminate the backlash in the mechanical transmission. The characteristics of the self-energized brake actuator in the infinite reinforcement region combined with a backlash in the mechanical transmission in the control loop will reduce the performance of the control loop, and it will increase the dissipated energy from responding to disturbances resulting from the backlash. In order to use the full advantage of the self-energization principle, it is important to use a control strategy that allow a fast and energy efficient way to control the brake actuator also in the infinite reinforcement region. As such, taking the backlash in the mechanical transmission into account in the control calculations is undesirable.
Moreover, it is well known that wheels may slip and lock up during severe braking or when braking on slippery road surfaces (e.g., wet or icy roads, wet leaves, etc.). This usually causes a long stopping distance and sometimes the vehicle may lose steering stability. The objectives of anti-lock brake systems (ABS) are to prevent wheels from locking up and to achieve minimum stopping distance while maintaining good steering stability (lateral stability) during braking. It is also well known that the wheel/road friction coefficient is a non-linear function of the slip, which is defined by the wheel velocity and the vehicle velocity. The ideal goal of the wheel brake controller is to regulate the wheel velocity such that a commanded or optimal slip is obtained, which corresponds to a commanded or maximum tire/road friction force. The commanded values may be provided from a superior vehicle control module, which controls multiple brakes (e.g., all brakes on the vehicle, all brakes on an axle, etc.).
A further deficiency of the prior art discussed above is that no adequate mechanism for achieving such control, either on a per wheel or a multi-wheel basis is disclosed.
It has been proposed to remedy the above problems in a commonly owned U.S. patent application entitled “System For Control Of Brake Actuator Based At Least In Part Upon Tire/Road Friction Force” filed simultaneously herewith, and naming several of the same inventors, by providing a vehicle brake system which includes a brake actuator associated with a wheel, the brake actuator having at least one moveable brake component, and a brake control system operative to determine a target rotational velocity of the wheel based at least in part on a relationship between an actual tire/road friction force and a target tire/road friction force, the brake control system determining a target position of the at least one moveable brake component based at least in part upon the target rotational velocity. The brake control system is operable to control actuation of the brake actuator based at least in part upon the target position of the at least one moveable brake component and at least in part based upon a current position of the at least one moveable brake component. The present invention is directed to a tire slip model which may be used in conjunction with such a brake control system in order to accurately predict the target tire/road friction force.
European Patent Application No. EP0949131 discloses control of the running behaviour of a four-wheeled vehicle using a mathematical tire model. However, the tire model disclosed therein treats tires together, rather than treating each tire individually. As such, the tire model disclosed in this reference is not particularly accurate or particularly quick and easy to calculate. U.S. Pat. No. 6,663,196 discloses a device for controlling wheel slip of individual wheels of a motor vehicle. However, no mention is made of any tire model which may be used for determining how the desired wheel slip is to be determined. U.S. Pat. No. 6,508,102 discloses a system and method for tire/road friction estimation. However, this reference does not disclose any type of tire model which takes into account an optimal tire slip; instead the reference is concerned with providing an indication of when tire slip is imminent.
What is desired, therefore, is a system for controlling application of an electronically controlled brake which is well-suited for controlling the application of an electromechanical brake with self-energizing characteristics, which facilitates the stability of a self-energizing brake for all possible disc/pad friction coefficients, which achieves the correct commanded torque/force on all brakes on the vehicle, which relies on sensor feedback to control application of the brake, which relies on sensor feedback indicative of other than the frictional force between the disc and pad, clamping force of the caliper or brake torque, which is not substantially deleteriously affected by a disc which has a thickness that varies during a revolution, which is not substantially deleteriously affected by ageing and corrosion of the sliding pins in a floating caliper design, which does not take into account backlash in the mechanical transmission in control calculations, which regulates wheel velocity such that a commanded or optimal slip that corresponds to a commanded or maximum tire/road friction force is obtained, and which regulates wheel velocity of multiple wheels such that whole vehicle control is optimized.